> there are a lot of .tmp files created - and the
> css-files are part of these .tmp-files.
tmp files are temporary internal processing. They should be replaced with the
actual file by completion.

> httrack <http://website> --robots=0 --keep-alive
> --extended-parsing
> +website/*.css +website/*.html +website/*.gif
> +website/*.jpg +website/*.png
> +website/*.swf  -mime:*/* +mime:text/html
> +mime:image/*  -*/print.html 
The default is to download everything from the starting site only. So those
first 6 filters do nothing. Then you say -mime:*/* which mean get nothing. So
your filters will only get html and images except print. No css files
are/should be expected.

Normally you want all those, no matter where they come from.
Drop the filters and just use the near flag -n instead of +*.css +*gif ...

Also some site do not like HTT browser ID. I always run with: --user-agent
"Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0)"